No correlation between germline mutation at repeat DNA and meiotic crossover in male mice exposed to...

Mutation Research 457 (2000) 79–91

No correlation between germline mutation at repeat DNA andmeiotic crossover in male mice exposed to X-rays or cisplatin

Ruth Barbera, Mark Plumba,b, Andrew G. Smithc, Carolina E. Cesara,Emma Boultonb, Alec J. Jeffreysa, Yuri E. Dubrovaa,d,∗

a Department of Genetics, University of Leicester, Leicester LE1 7RH, UKb Medical Research Council Radiation and Genome Stability Unit, Harwell, Oxon OX11 ORD, UK

c Medical Research Council Toxicology Unit, University of Leicester, Leicester, LE1 9HN, UKd N.I. Vavilov Institute of General Genetics, Russian Academy of Sciences, Moscow B-333, Russia

Received 12 July 2000; received in revised form 5 September 2000; accepted 8 September 2000

Abstract

To test the hypothesis that mouse germline expanded simple tandem repeat (ESTR) mutations are associated with recom-bination events during spermatogenesis, crossover frequencies were compared with germline mutation rates at ESTR loci inmale mice acutely exposed to 1 Gy of X-rays or to 10 mg/kg of the anticancer drug cisplatin. Ionising radiation resulted ina highly significant 2.7–3.6-fold increase in ESTR mutation rate in males mated 4, 5 and 6 weeks after exposure, but not 3weeks after exposure. In contrast, irradiation had no effect on meiotic crossover frequencies assayed on six chromosomesusing 25 polymorphic microsatellite loci spaced at approximately 20 cM intervals and covering 421 cM of the mouse genome.Paternal exposure to cisplatin did not affect either ESTR mutation rates or crossover frequencies, despite a report that cisplatincan increase crossover frequency in mice.

Correlation analysis did not reveal any associations between the paternal ESTR mutation rate and crossover frequencyin unexposed males and in those exposed to X-rays or cisplatin. This study does not, therefore, support the hypothesisthat mutation induction at mouse ESTR loci results from a general genome-wide increase in meiotic recombination rate.© 2000 Elsevier Science B.V. All rights reserved.

Keywords:Crossing over; Expanded simple tandem repeat loci; Minisatellite loci; Microsatellite loci; Radiation; Cisplatin; Mouse

1. Introduction

Analysis of mutation induction at mouse expandedsimple tandem repeat (ESTR)1 loci has shown that

∗ Corresponding author. Tel.:+44-116-252-5654;fax: +44-116-252-3378.E-mail address:[email protected] (Y.E. Dubrova).

1 ESTR loci were originally termed minisatellites but have re-cently been renamed to distinguish them from the much morestable true minisatellites in the mouse genome [4,5].

exposure to ionising radiation results in a lineardose-response in germline ESTR mutation rates de-tectable at doses substantially lower than can be moni-tored by standard genetic techniques [1–3]. Moreover,in sharp contrast to previously used genetic systems,the very high rate of spontaneous mutation alteringallele length (repeat copy number) allows germlinemutation induction to be detected in very small popu-lation samples. However, the mechanism of mutationinduction at mouse ESTR loci remains unknown.Given the very high frequency of radiation-induced

0027-5107/00/$ – see front matter © 2000 Elsevier Science B.V. All rights reserved.PII: S0027-5107(00)00130-5

80 R. Barber et al. / Mutation Research 457 (2000) 79–91

mutation, it is most unlikely that these small ge-nomic loci (less than 12 kb) are the direct targetsof irradiation [2,3,6,7]. Instead, it appears that mu-tation induction is a two-stage process involving aradiation-induced cellular signal that subsequentlytriggers instability at ESTRs. The persistence of in-stability into the germline of offspring of irradiatedmice, resulting in transgenerational mutagenesis [8],provides further evidence for a non-targeted processthat may involve meiotic recombination, DNA repairor other unknown processes.

Detailed information on the processes that drivethis type of repeat DNA instability only exists forhuman minisatellites, and may provide clues to themechanisms of mutation induction at mouse ESTRloci. All GC-rich minisatellites studied mutate bya recombination-based process specifically in thegermline that usually results in the inter-allelic trans-fer of blocks of repeat units [9]. Such conversionsprobably arise from meiotic double strand breaks inthe repeat array that initiate recombination events,including not only conversions but also crossovers[10,11]. For minisatellite MS32 at least, crossoveractivity is not restricted to the repeat array but ex-tends into flanking DNA on one side of the array,defining an intense and highly localised hotspotfor meiotic crossover centred upstream of the min-isatellite [12]. Current evidence suggests that it isthis hotspot that drives minisatellite instability andthat crossover and conversion can originate fromthe same initiation complex [10,12]. If meiotic re-combination also plays a role in mutation at mouseESTR loci, then it is possible that radiation-inducedmutation may arise from a genome-wide in-crease in meiotic recombination frequencies. Ifso, then similar increases in ESTR mutation rateand crossover frequency should occur in exposedanimals.

Little information exists on the susceptibility ofmeiotic recombination to mutagens, with most stud-ies focusing on the induction of somatic recombi-nation by ionising radiation and different chemicals[13]. Crossing-over in germ cells has been analysedin Drosophilamales which show a complete suppres-sion of meiotic recombination and increases in recom-bination rate after exposure to ionising radiation andchemical mutagens, therefore, demonstrate inductionin mitotic rather than meiotic cells [14,15]. Similarly, a

recently developed transgenic mouse system designedto detect recombination events in the germline [16] ismost likely detecting gene conversions and crossoversin mitotic germ cells rather than at meiosis. Anotherattempt to explore the morphology of the synaptone-mal complex as an indicator of chromosome damagefailed to address the issue of crossover induction atmeiosis [17].

For most of the existing mammalian experimentalmodels, very few polymorphic linked genes with aclearly visible phenotypic effect have been identifiedand this complicates the analysis of meiotic recombi-nation. Thus, the recent study showing reduced fre-quency of crossing-over between thep andTyr(c) lociin male mice exposed at the early meiotic stages to thetopoisomerase-II inhibitor etoposide was restricted toa relatively small 14 cM interval of chromosome seven[18]. The availability of detailed microsatellite-basedlinkage maps [19] greatly facilitates the analysis ofmeiotic crossover rates over much larger genomic in-tervals and has recently been used to show that theanticancer drug cisplatin increases meiotic crossoverrates in mice [20].

As radiation-induced ESTR germline mutations andcisplatin-induced increases in crossover rates occur atspecific stages of mouse spermatogenesis [2,20], thisraises the possibility that they are related. We, there-fore, now use a panel of microsatellite loci coveringmost of six mouse chromosomes (28% of the genomein total), together with a set of hypervariable ESTRloci, to directly test the hypothesis that ESTR insta-bility results from a genome-wide increase in meioticcrossover rates in male mice exposed to X-rays or cis-platin.

2. Materials and methods

2.1. Mouse breeding and irradiation

F1 hybrid males (C57BL/6 × CBA/H) andCBA/H females were from the Harwell colony. Tennon-treated males (8–12-week-old) were crossed tountreated females to obtain control offspring (Table 1).Two weeks later, the same 10 males were exposed to1 Gy X-rays delivered at 0.5 Gy/min (250 kV constantpotential, HLV 1.2 mm Cu) and mated to unexposedCBA/H females 3, 4, 5 and 6 weeks post-irradiation.

R. Barber et al. / Mutation Research 457 (2000) 79–91 81

Table 1Summary of ESTR mutation data

Exposurestage

No. oflitters

No. ofoffspring

Number of mutations scored by different probesa x2

Grandtotal

MMS10 Ms6-hm Hm-2 MS6-hm+Hm-2

d.f.=1b Probabilityc

Paternal Maternal Paternal Maternal Paternal Maternal

X-raysControl 19 119 56 (46) 22 (21) 11 (7) 11 (10) 6 (3) 6 (5) 17 (10) 17 (15) 0.00 –3 weeks 8 57 30 (29) 10 (10) 8 (7) 6 (6) 3 (3) 3 (3) 11 (10) 9 (9) 0.20 0.65474 weeks 9 55 58 (54) 28 (27) 16 (14) 4 (3) 7 (7) 3 (3) 23 (21) 7 (6) 8.850.00295 weeks 5 39 38 (33) 18 (17) 11 (8) 5 (4) 4 (4) 0 15 (12) 5 (4) 5.100.02396 weeks 7 51 74 (70) 35 (34) 17 (16) 6 (5) 9 (8) 7 (7) 26 (24) 13 (12) 4.360.0368

CisplatinControl 14 71 44 (37) 19 (19) 8 (7) 4 (4) 6 (6) 7 (5) 14 (13) 11 (9) 0.35 0.55413 weeks 10 54 24 (18) 12 (11) 6 (6) 1 (1) 1(1) 4 (3) 7 (7) 5 (4) 0.32 0.57164 weeks 8 37 20 (18) 9 (9) 2 (2) 5 (4) 3 (3) 1 (1) 5 (5) 6 (5) 0.09 0.76425 weeks 9 51 27 (22) 13 (13) 5 (4) 6 (3) 1 (1) 2 (2) 6 (5) 8 (5) 0.28 0.59766 weeks 11 62 41 (36) 17 (17) 9 (8) 7 (5) 5 (5) 3 (3) 14 (13) 10 (8) 0.66 0.4166

a Number of independent mutations are given in parentheses.b x2-test for equal number of paternal and maternal mutations.c Probability for x2-test, statistically significant values are in bold.

Animal procedures were carried out under guidanceissued by the MRC in ‘Responsibility in the use ofanimals for medical research’ (July 1993) and underHome Office project licences No. PPL 30/689 and30/1272.

2.2. Mouse breeding and cisplatin treatment

F1 hybrid males (C57BL/6J × CBA/Ca) andCBA/Ca females were purchased from Harlan Ltd.,UK. Ten non-treated males (8-week-old) were crossedto untreated females to produce control offspring(Table 1). One week later they were given a singleintraperitoneal dose of cisplatin (Sigma, 10 mg/kg,at 1 mg/ml in 0.9% NaCl). These males were matedto untreated CBA/Ca females 3, 4, 5 and 6 weekspost-treatment. In a separate study, males injectedwith cisplatin were culled after 1 week and the testesremoved, fixed in buffered formalin and retained forhistological examination. Histological examinationshowed mild effects on the seminiferous epithelium,apparently in spermatogonia (data not shown), consis-tent with those reported previously [21] and confirm-ing that our dosing regime was delivering cisplatinto the testes. All animal procedures were carried outunder guidance issued by the MRC in ‘Responsibil-ity in the use of animals for medical research’ (July

1993) and under Home Office project licence No.PPL 80/1353.

2.3. DNA isolation and ESTR typing

DNA was prepared from tails using phenol–chloroform extraction [22]. DNA (5mg) was digestedto completion withAluI, electrophoresed through a40 cm long 0.8% (w/v) agarose gel (SeaKem typeLE, FMC) in 1 × TBE buffer (89 mM Tris-borate,pH 8.3, 2 mM EDTA), transferred to a nylon mem-brane (Magna Nylon, MSI Osmonics) and hy-bridised to 32P-labelled probes as described else-where [23]. ESTR mutations were scored using therodent-specific multilocus ESTR probe MMS10 [4]and two mouse-specific hypervariable single-locusESTR probes Ms6-hm and Hm-2 [24,25] over thewell-resolved region between 2.5 and 22 kb. Muta-tions were scored as a novel size band in offspringnot present in either parent [1–4,8].

2.4. Microsatellite typing

PCR primers for 25 mouse microsatellites wereobtained from Research Genetics and PE AppliedBiosystems. Chromosome assignments of microsatel-


Table 2Chromosome assignment of studied microsatellite loci (data from[25])

Locus Position (cM) Distance (cM)a

Chromosome 1D1Mit231 12.0 –D1Mit156 32.8 20.8D1Mit187 62.0 29.2D1Mit15 87.9 25.9D1Mit17 106.3 18.4

Total – 94.3 (84.2%)

Chromosome 2D2Mit433 31.7 –D2Mit420 56.4 22.9D2Mit412 78.7 24.1D2Mit265 105.0 26.3

Total – 73.3 (64.3%)


Total – 71.1 (74.8%)

Chromosome 4D4Mit149 0 –D4MIT286 14.5 14.5D4Mit308 57.4 42.9D4Mit54 66.0 8.6

Total – 66.0 (76.8%)


Total – 51.0 (60.7%)


Total – 66.0 (82.5%)

Grand total 421.7 (28.1%)

a The percentage of chromosome/genome flanked by mi-crosatellite loci is given in brackets.

lite loci [26] are given in Table 2. Mouse genomicDNAs were amplified in 15ml reactions using 0.2 mMdNTPs, 0.2mM of each primer, 25 or 75 ng of tem-plate DNA, 0.5 units of Taq polymerase (Advanced

Biotechnologies), 1×PCR buffer (Advanced Biotech-nologies) and 1–3 mM MgCl2. Amplifications wereperformed in thin-walled 96 well-plates on a MJ DNAEngine PTC 225. After denaturing at 95◦C for 5 min,PCR reactions were cycled at 95◦C for 30 s, 52–60◦Cfor 30 s, and 72◦C for 1 min for 28 cycles, endingwith a 10 min incubation at 72◦C. 10ml of PCR prod-ucts were electrophoresed through a 4% agarose gel(MetaPhor agarose, FMC-Bioproducts) in 0.5 × TBEbuffer, containing 0.5mg/ml of ethidium bromide. Toensure the accuracy of microsatellite typing, all PCRreactions were run in duplicate, and any apparentdouble recombinants were additionally retyped.

3. Results

3.1. Experimental design

Crossover frequencies and ESTR mutation rateswere measured in five groups of offspring conceivedeither before or 3, 4, 5 and 6 weeks after acute ex-posure to X-rays or cisplatin injection. This matingscheme allowed the detection of mutation induction atthe relevant stages of mouse spermatogenesis, diaki-nesis and meiotic divisions (3 weeks post-treatment),mid-late pachytene (4 weeks), type B spermatogo-nia (5 weeks) and type As spermatogonia (6 weeks)[27,28]. Crossing-over occurs sometime during thepachytene/diplotene phases approximately 22–30days after treatment [27,28]. The heterozygous F1hybrids were backcrossed with homozygous CBA fe-males to permit the detection of single and multiplecrossover events over the 28% of the mouse genomecovered by the microsatellite loci selected (Table 2).Since at least one meiotic recombination event perchromosome is required to produce a fertile gamete,any induction of recombination events would appearas an increase in double (or multiple) recombinationevents per chromosome. However, our ability to dis-tinguish between single or double/multiple crossoverevents was limited by the constraints imposed by thedensity of microsatellite loci employed.

3.2. ESTR mutation and crossover frequency inirradiated males

Mutation data are summarised in Table 1. Thesedata also address the issue of mutation clustering,


Fig. 1. ESTR mutation rates and crossover frequencies in male mice following irradiation by 1 Gy of X-rays: a, b paternal ESTR mutationrate (a aggregated data, b mutation rate in each male); c, d crossover frequencies (c aggregated data, d crossover frequency in each male).95% confidence intervals for mutation rate and crossover frequency are given in a and c.

when apparently identical mutants are shared by morethan one offspring in a litter. We have previouslyshown that mutation clustering at mouse minisatellitesMs6-hm and Hm-2 usually results from germ-linemutational mosaicism [24,25]. Table 1, therefore,shows the number of all mutations, together with thenumber of different mutations with clusters countedas a single mutational event. In the following anal-ysis, we use all new mutant alleles, irrespective ofwhether they show clustering. However, reduction ofall shared mutations to a single mutation caused onlya minor decrease in estimated mutation rates and didnot affect the magnitude of mutation rate differencesseen between the various exposure regimes (data notshown).

Frequencies of ESTR mutation are given in Table 3.Offspring conceived 3 weeks after exposure showedno increase in ESTR mutation rates (Fig. 1 (a)). In

contrast, a highly significant 2.7–3.6-fold increase inpaternal mutation rate was found in offspring con-ceived 4, 5 and 6 weeks after treatment. In contrast,paternal irradiation did not affect the mutation rate ofalleles transmitted from the unexposed mothers (Ta-ble 3). Similar results were obtained with the indepen-dent multi-locus probe MMS10, although, the parentalorigin of mutants could not be determined (Table 3).

Crossover frequencies in the same irradiated malesremained unchanged over the whole period of expo-sure and were similar to those seen in the same micebefore irradiation (Table 4, Fig. 1(c)). There was noevidence for increased crossover frequencies in any ofthe six chromosomes studied (Table 4) nor in any ofthe exposed males (Fig. 1(d)). Furthermore, irradia-tion had no apparent effect on the frequency of doublecrossovers (Table 4). Finally, correlation analysis didnot reveal any significant associations between pater-

84R

.B

arb

er

et

al./M

uta

tion

Re

sea

rch4

57

(20

00

)7

9–

91

R.

Ba

rbe

re

ta

l./Mu

tatio

nR

ese

arch

45

7(2

00

0)

79

–9

185


Fig. 2. ESTR mutation rates and crossover frequencies in male mice exposed to 10 mg/kg of cisplatin: a, b paternal ESTR mutation rate (aaggregated data, b mutation rate in each male); c, d crossover frequencies (c aggregated data, d crossover frequency in each male). 95%confidence intervals for mutation rate and crossover frequency are given in a and c.

nal ESTR mutation rates and crossover frequencies inthe control and exposed males (Fig. 3(a)).

The data summarised in Table 4 also allowed us toestimate the statistical power of the crossover analy-sis. Using the maximum standard error (26.5) of thecrossover frequency within one experimental group (5weeks after exposure) and the grand total for the con-trol group, we estimate that a≥ 15% increase in over-all crossover frequency in any exposed group wouldbe detected with 95% confidence. However, we donot have sufficient statistical power to detect changesin crossover frequencies within each∼20 cM interval.This also applies to the male mice exposed to cisplatin.

3.3. ESTR mutation and crossover frequency incisplatin-treated males

The anticancer drug cisplatin was chosen, givenevidence that it induces meiotic crossover in mice

[20] and might, therefore, induce ESTR instabil-ity if mutation at these loci involves recombina-tion. However, the frequency of ESTR mutationin all offspring conceived after paternal exposureto cisplatin was similar to that in mice conceivedbefore exposure (Tables 1 and 3, Fig. 2(a)). Noevidence for mutation induction was found forany of the ESTR loci (Table 3), nor for any ofthe exposed males (Fig. 2(b)). Similarly, crossoverfrequencies in the cisplatin-treated males did notdiffer from those before treatment, showing onlyrandom fluctuations in each of the six chromo-somes studied and in each of the exposed malesover the whole period of exposure (Table 5,Fig. 2(c), (d)). As with irradiated males, againthere was no significant correlation between pater-nal ESTR mutation rates and crossover frequenciesin male mice before and after cisplatin treatment(Fig. 3(b)).

R.

Ba

rbe

re

ta

l./Mu

tatio

nR

ese

arch

45

7(2

00

0)

79

–9

187


Fig. 3. Correlation between crossover frequencies and paternal ESTR mutation rates in male mice before and after exposure to X-rays;(a) Kendall’sτ = 0.0920,P = 0.7113 andτ = −0.0734,P = 0.5911 for males before and after exposure, respectively) or cisplatin (b)Kendall’s τ = −0.0698,P = 0.7786 andτ = −0.1003,P = 0.4279 for males before and after exposure, respectively).

4. Discussion

The purpose of this study was to establish whetherelevated germline mutation at mouse ESTR loci couldbe attributed to a genome-wide induction of meioticcrossover. To address this issue, we exposed male miceto X-rays, which are known to dramatically increaseESTR germline mutation rates [2] but have an un-known effect on meiotic crossover and to cisplatin,which is reported to increase meiotic crossover rates

in mice [20] but has not been studied with respect toESTR mutation. If ESTR mutation induction is drivenby meiotic recombination, then both X-ray and cis-platin treatment should give a proportionate increaseboth in ESTR mutation rates and meiotic crossoverfrequencies in exposed mice.

Analysis of irradiated males showed clear evidencefor mutation induction at ESTR loci but no effects onmeiotic crossover frequencies nor any correlation be-tween crossover rates and ESTR mutation in control


and irradiated males, suggesting that ESTR mutationinduction is not due to an increase in meiotic crossingover. While radiation can induce double strand breaks(DSBs) that may serve to initiate recombination, suchDSBs in pre-meiotic cells are likely to be repairedby non-homologous end-joining and recombination,without crossover between homologous chromosomes[29]. Alternatively, cells containing DSBs may be lostby apoptosis. In theory, DSBs induced by radiationat the crossover-proficient stages of spermatogenesiscould lead to an increase in meiotic crossover rates.While our experimental design included these stagesof mouse spermatogenesis (mice bred 4 weeks afterexposure), no evidence for crossover induction wasfound. However, the exposed population of germ cellsat 4 weeks is highly heterogeneous, and it remainspossible that strict targeting of mouse germ cells atvery specific stages of spermatogenesis may result inelevated crossover frequency, though the timing of in-duction may be very short and difficult to analyse.

We have previously shown that ionising radiationinduces ESTR mutation in exposed pre-meiotic sper-matogonia (6 weeks post-treatment) and stem cells (10and 20 weeks) but not 3 weeks after exposure [2]. Thepresent study extends the analysis to two additionalstages of mouse spermatogenesis (4 and 5 weeks af-ter exposure), demonstrating a significant increase inpaternal mutation rate following exposure of cells atmid and early pachytene and type B spermatogoniawhich is comparable to that in type As spermatogonia(Table 3, probabilities of difference between stages,0.4 < P < 0.95), and confirms the lack of induction 3weeks after exposure [2]. Together, these results sug-gest that radiation-induced ESTR mutations can occurat all stages of spermatogenesis prior to metaphase I.

Unfortunately, spontaneous mutation processes atESTR loci in the mouse germline remain unknown.Unlike human GC-rich minisatellites which mutate al-most exclusively at meiosis [9–12], mouse ESTR locialso show significant somatic instability, particularlyin the first few cell divisions following fertilisation,resulting in the frequent appearance of germline andsomatic mutational mosaicism [24,25]. Similarly, ir-radiation can result in ESTR instability persisting intothe germline of the offspring of exposed mice [8],again indicating that mutation can arise in diploid stemcells and is not restricted to meiosis. Our data sug-gest that non-targeted radiation-induced mutation at

ESTR loci [2,3,6,7] can occur in all pre-meiotic sper-matogenic cells, but not in cells that have undergonecrossing-over.

The anticancer drug cisplatin is a powerful DNAcross-linking agent capable of forming a variety ofDNA adducts, mostly 1, 2-intrastrand cross-links[30]. Exposure of somatic cells to this drug inducesmutations at protein-coding genes, as well as re-combination events, sister-chromatid exchanges andchromosome loss [31]. It also induces chromosomeaberrations in mouse spermatocytes [32] with a dou-bling dose 0.5 mg/kg [33] and in spermatogonial stemcells [34]. However, there is no evidence that cisplatininduces germline mutations (dominant lethals andspecific locus mutations) in male mice exposed to rel-atively high levels of cisplatin (10 mg/kg) at differentstages of spermatogenesis [35–37]. Similarly, we findno evidence that the same level of cisplatin inducesESTR mutations at any of the stages of germ-celldevelopment analysed. The reasons for resistance ofthe mouse germline to cisplatin mutagenesis remainunclear. It could include the existence of very efficientDNA repair systems in pre-meiotic germ cells capableof high-fidelity removal of cisplatin-induced damage,as well as DNA damage triggering cell loss by apop-tosis, plus cytotoxic effects on pre- and post-meioticcells [34]. In the latter case, this could lead to theselective loss of sperm carrying cisplatin-inducedDNA damage that otherwise could subsequently leadto the appearance of induced mutations. However,the negative results for post-meiotic stages of sper-matogenesis [35–37], where DNA repair is greatlysuppressed, plus the very high toxicity of cisplatinon mouse spermatogenic cells [34], together indicatethat the lack of mutation induction may largely beattributed to apoptosis and direct cell-killing effects.

Surprisingly, cisplatin also had no effect on mei-otic crossover frequencies on any of the six chro-mosomes analysed in any male studied 3, 4, 5 and6 weeks after exposure (Table 5, Fig. 2(c), (d)).These findings contrast with recent evidence forcisplatin-induced crossovers in male mice mated 4weeks post-treatment [20]. Both the study in [20] andthe present analysis used the same dose of cisplatinand the same microsatellite-based approach to scor-ing crossovers, including crossovers on chromosome10. This discrepancy cannot be attributed to the sta-tistical power of our study. The 1.8-fold increase in


crossover frequency in mice exposed at this stage ofspermatogenesis reported in [20] should have resultedin a total crossover frequency of 1.8 × 337 = 606%in the present study (Table 5), whereas the observedfrequency at this stage was 362± 27% (4 weeks af-ter exposure, Table 5). The difference between thesetwo values corresponds to 8.9 standard error unitsand the probability of this is well below 10−8. Thecorresponding estimates for chromosome 10 anal-ysed in [20] and in our study yielded a probability2 × 10−5. The discrepancy may instead be due to thedifferent F1 genotypes being used in the two studies(C57BL/6J×DBA/2J in [20], C57BL/6J×CBA/Cain our investigation). While strain CBA was devel-oped in 1920 from a cross between a Bagg albinofemale and a DBA male [38] and should still sharesome common genetic background with DBA, it re-mains possible that F1 mice derived from these twostrains differ in their response to cisplatin, if correct,this remarkable strain difference would merit furtherinvestigation.

In conclusion, we present evidence that acute ex-posure to X-rays results in a significant increase inthe ESTR mutation rate at all stages of spermato-genesis prior to metaphase I, whereas treatment ofmale mice with the anticancer drug cisplatin does notaffect germline mutation rate at these loci. Similarly,previous analysis of protein-coding genes revealedthe high mutagenic potential of X-rays and lack ofgermline mutation induction in cisplatin-treated malemice [28,35–37]. These parallels between ESTR andprotein-coding gene mutations suggest a possiblemechanistic connection between these two classesof mutation in diploid cells which, if correct, wouldstrengthen the relevance of using ESTR loci to mon-itor germline mutation. Finally, neither X-rays norcisplatin affected meiotic crossover frequencies, indi-cating that mutation induction at ESTR loci cannotbe attributed to a genome-wide increase in meioticrecombination rates.

Acknowledgements

We thank, Dr. M. Tucker for pathological expertise.This work was supported by grants to Y.E.D. and A.J.J.from the Wellcome Trust, by grants to A.J.J. from theMedical Research Council and the Royal Society and

by a grant to M.P. from the Leukaemia Research Fund.A.J.J. is also supported in part by an InternationalResearch Scholars Award from the Howard HughesMedical Institute. C.E.C. work was supported by grantfrom the Nuffield Foundation.

References

[1] Y.E. Dubrova, A.J. Jeffreys, A.M. Malashenko, Mouseminisatellite mutations induced by ionizing radiation, NatureGenet. 5 (1993) 92–94.

[2] Y.E. Dubrova, M. Plumb, J. Brown, J. Fennelly, P. Bois,D. Goodhead, A. J Jeffreys, Stage specificity, dose-responseand doubling dose for mouse minisatellite germline mutationinduced by acute radiation, Proc. Natl. Acad. Sci. USA 95(1998) 6251–6255.

[3] Y.E. Dubrova, M. Plumb, J. Brown, E. Boulton, D. Goodhead,A.J. Jeffreys, Induction of minisatellite mutation in the mousegermline by low-dose chronic exposure tog-radiation andfission neutrons, Mutat. Res. 453 (2000) 67–75.

[4] P. Bois, J. Williamson, J. Brown, Y.E. Dubrova, A.J. Jeffreys,A novel unstable mouse VNTR family expanded from SINEB1 element, Genomics 49 (1998) 122–128.

[5] P. Bois, J.D.H. Stead, S. Bakshi, J. Williamson, R.Neumann, B. Moghadaszadeh, A.J. Jeffreys, Isolation andcharacterization of mouse minisatellites, Genomics 50 (1998)317–330.

[6] Y.J. Fan, Z. Wang, S. Sadamoto, Y. Ninomiya, N.Kotomura, K. Kamiya, K. Dohi, R. Kominami, O. Niwa,Dose-response of radiation induction of a germline mutationat a hypervariable mouse minisatellite locus, Int. J. Radiat.Biol. 68 (1995) 177–183.

[7] Y.E. Dubrova, M. Plumb, J. Brown, A.J. Jeffreys,Radiation-induced germline instability at minisatellite loci,Int. J. Radiat. Biol. 74 (1998) 689–696.

[8] Y.E. Dubrova, M. Plumb, B. Gutierrez, E. Boulton, A.J.Jeffreys, Transgenerational mutation by radiation, Nature 405(2000) 37.

[9] A.J. Jeffreys, K. Tamaki, A. MacLeod, D.G. Monckton, D.L.Neil, A.J.L. Armour, Complex gene conversion events ingermline mutation at human minisatellites, Nature Genetics6 (1994) 136–145.

[10] A.J. Jeffreys, D.L. Neil, R. Neumann, Repeat instabilityat human minisatellites arising from meiotic recombination,EMBO J. 17 (1998) 4147–4157.

[11] J. Buard, A.C. Jeffreys, A.J. Shone, Meiotic recombinationand flanking marker exchange at the highly unstable humanminisatellite CEB1 (D2S90), Am. J. Hum. Genet. 67 (2000)333–344.

[12] A.J. Jeffreys, J. Murray, R. Neumann, High-resolutionmapping of crossovers in human sperm defines aminisatellite-associated recombination hotspot, Molec. Cell 2(1998) 267–273.

[13] G.R. Hoffmann, Induction of genetic recombination:consequences and model systems, Environ. Mol. Mutagen.23 (1994) 59–66.


[14] L.E. Schacht, The time of X-ray induction of crossovers andtranslocations inDrosophila melanogastermales, Genetics43 (1958) 665–678.

[15] F.H. Sobels, H. van Steenis, Chemical induction ofcrossing-over inDrosophilamales, Nature 179 (1957) 29–31.

[16] J.R. Murti, K.J. Schimenti, J.C. Schimenti, A recombination-based transgenic mouse system for genotoxicity testing,Mutat. Res. 307 (1994) 583–595.

[17] M.J. Moses, P. Poorman-Allen, J.H. Tepperberg, J.B. Gibson,L.C. Backer, J.W. Allen, The synaptonemal complex as anindicator of induced chromosome damage, in: J.W. Allen,B.B. Bridges, M.F. Lyon, M.J. Moses, L.B. Russell (Eds.),Biology of Mammalian Cell Mutagenesis, Cold Spring HarborLaboratory Press, 1990, pp. 133–151.

[18] L.B. Russell, P.R. Hunsicker, A.M. Hack, T. Ashley, Effectof the topoisomerase-II inhibitor etoposide on meioticrecombination in male mice, Mutat. Res. 464 (2000) 201–212.

[19] M. Rhodes, R. Straw et al., A high-resolution microsatellitemap of the mouse genome. Genome Res. 8 (1998) 531-542.

[20] W.H. Hanneman, M.E. Legare, S. Sweeney, J.C. Schimenti,Cisplatin increases meiotic crossing-over in mice, Proc. Natl.Acad. Sci. USA 94 (1993) 8681–8685.

[21] P. Kopfer-Maier, Effects of carboplatin on the testis. Ahistological study, Cancer Chemother. Pharmacol. 29 (1992)227–235.

[22] N.D. Allen, S.C. Barton, M.A.H. Surani, W. Reik, MammalianDevelopment: a Practical Approach, IRL Press, Oxford, 1987.

[23] Z. Wong, V. Wilson, A.J. Jeffreys, S.L. Thein, Cloning aselected fragment from human DNA “fingerprint” isolationof an extremely polymorphic minisatellites, Nucl. Acids Res.14 (1986) 4605–4616.

[24] R. Kelly, G. Bulfield, A. Collick, M. Gibbs, A.J.Jeffreys, Characterization of a highly unstable mouseminisatellite locus: evidence for somatic mutation during earlydevelopment, Genomics 5 (1989) 844–856.

[25] M. Gibbs, A. Collick, R. Kelly, A.J. Jeffreys, Atetranucleotide repeat mouse minisatellite displayingsubstantial somatic instability during early preimplatationdevelopment. Genomics 17 (1993) 121–128.

[26] J.A. Blake, J.T. Eppig, J.E. Richardson, M.T. Davisson,The Mouse Genome Database Group, The Mouse GenomeDatabase (MGD): expanding genetic and genomic resources

for the laboratory mouse, Nucl. Acids Res. 28 (2000) 108–111.

[27] E.F. Oakberg, Germ cell toxicity: significance in genetic andfertility effects of radiation and chemicals, Environ. Sci. Res.31 (1984) 549–590.

[28] A.G. Searle, Mutation induction in mice, Adv. Radiat. Biol.4 (1974) 131–207.

[29] D.T. Weaver, Regulation and repair of double-strandDNA breaks, Crit. Rev. Euk. Gene Exp. 64 (1996) 345–375.

[30] D.B. Zamble, S.J. Lippard, Cisplatin and DNA repair incancer chemotherapy, Trends Biochem. Sci. 20 (1995) 435–439.

[31] B.J.S. Sanderson, L.R. Ferguson, W.A. Denny, Mutagenic andcarcinogenic properties of platinum-based anticancer drugs,Mutat. Res. 355 (1996) 59–70.

[32] I.-D. Adler, A. El-Tarras, Clastogenic effects ofcis-diamminedichloroplatinum (II). Induction of chromosomalaberrations in primary spermatocytes and spermatogonial stemcells of mice, Mutat. Res. 243 (1990) 173–178.

[33] I.-D. Adler, A. El-Tarras, Clastogenic effects ofcis-diamminedichloroplatinum (I). Induction of chromosomalaberrations in somatic and germinal cells of mice, Mutat.Res. 211 (1989) 131–137.

[34] M.L. Meistrich, M. Finch, M.F. da Cunha, U. Hacker, W.W.Au, Damaging effects of 14 chemotherapeutic drugs on mousetestis cells, Cancer Res. 42 (1982) 122–131.

[35] M.A. Katoh, K.T. Cain, L.A. Hughes, L.B. Foxworth, J.B.Bishop, W.M. Generoso, Female-specific dominant lethaleffects in mice, Mutat. Res. 230 (1990) 205–217.

[36] L.B. Russell, W.L. Russell, E.M. Rinchik, P.R. Hunsicker,Factors affecting the nature of induced mutations, in:J.W. Allen, B.A. Bridges, M.F. Lyon, L.B. Russell (Eds.),Biology of Mammalian Cell Mutagenesis, Cold Spring HarborLaboratory Press, 1990, pp. 271–289.

[37] K.L. Witt, J.B. Bishop, Mutagenicity of anticancer drugs inmammalian germ cells, Mutat. Res. 355 (1996) 209–234.

[38] M. Festing, Origins and characteristics of inbred strainsof mice, in: M.F. Lyon, S. Rastan, S.D.M. Brown (Eds.),Genetic Variants and Strains of the Laboratory Mouse,Oxford University Press, Oxford, Vol. 2, 1996, pp. 1537–1576.

No correlation between germline mutation at repeat DNA and meiotic crossover in male mice exposed to...

Documents

Transcript of No correlation between germline mutation at repeat DNA and meiotic crossover in male mice exposed to...